Review of past on-orbit collisions among cataloged objects and examination of the catastrophic fragmentation concept☆

Abstract By mid-2013, four accidental hypervelocity collisions among cataloged objects have been recorded in low Earth orbit. Three were debris–intact impacts, generating very few cataloged fragments (less than 10, in total), while one involved the collision between two intact spacecraft, one of which maneuverable, with the consequent production of more than 2000 cataloged fragments. In order to evaluate the past, current and future average collision rates, the results of debris evolution simulations carried out in 1999–2000 were carefully re-analyzed and the corresponding collision rates, among objects larger than 10 cm, were estimated from 1980 to 2040. The value predicted for mid-2013, 0.240 events per year, was further validated by analyzing the current unclassified debris catalog and assessing its qualitative and quantitative incompleteness. It was also found that the intact–intact average collision rate is approximately 30% of the total, while debris–debris impacts account for less than 2%, the majority of the events involving debris–intact collisions. The potential benefits of a wide adoption of collision avoidance practices for maneuverable spacecraft was analyzed as well. If applied to 50% of the current operational spacecraft below 2000 km, it would lead to an 11% reduction of the overall collision rate and to a 16% reduction of the intact–intact collision rate. The model estimates from 1980 to mid-2013 were then used to compute the probability of collision occurrence as a function of the time interval considered. It was found that the number of accidental collisions observed so far is in extremely good agreement with the model predictions, corresponding to the maximum likelihood of the probability distributions. Moreover, even the proportion between debris–intact and intact–intact events reflects well what should be expected. Concerning the projections up to 2040, the probability of avoiding further collisions is less than 1%, while there is a probability of more than 50% of having 8 additional collisions, or more. Focusing the attention on the intact–intact events, the probability of avoiding another intact–intact collision is around 8%. At least 1 further collision will have therefore a probability to occur higher than 90%, 2 or more collisions will have a probability higher than 70%, and 3 or more collisions will have a probability higher than 45%, even though the broad adoption of collision avoidance for maneuverable spacecraft might reduce these figures. Finally, the concept of catastrophic collision was re-analyzed, taking into account the observational record and the basic assumptions usually adopted by the long-term debris evolution models. Potential sources of collision fragments over- and underestimation were identified and discussed, coming to the preliminary conclusion that after one century the number of collision fragments currently predicted by the evolutionary models might be overestimated.